RESEARCH ARTICLE
Potential of pest regulation by insectivorous birds in Mediterranean woody crops Jose´ M. Rey Benayas1,2*, Jorge Meltzer2, Daniel de las Heras-Bravo1¤, Luis Cayuela2,3 1 Life Sciences Department-Ecology Unit, Alcala University, Alcala´ de Henares, Spain, 2 International Foundation for Ecosystem Restoration, Madrid, Spain, 3 Department of Biology, Geology, Physics and Inorganic Chemistry, Rey Juan Carlos University, Mo´stoles, Spain
a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
OPEN ACCESS Citation: Rey Benayas JM, Meltzer J, de las HerasBravo D, Cayuela L (2017) Potential of pest regulation by insectivorous birds in Mediterranean woody crops. PLoS ONE 12(9): e0180702. https:// doi.org/10.1371/journal.pone.0180702 Editor: Tim A. Mousseau, University of South Carolina, UNITED STATES Received: January 31, 2017 Accepted: June 20, 2017 Published: September 6, 2017 Copyright: © 2017 Rey Benayas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This research was funded by the Spanish Ministry of Economy and Competitivity (grant numbers CGL2010-18312 and CGL2014-53308-P to JMRB), the Spanish Ministry of Food, Agriculture and the Environment (calls to foundations numbers ARM/934/2011, AAA/2286/ 2012, AAA/1284/2014, and AAA/1770/2015 to JMRB), the government of Madrid (grant numbers S2009AMB-1783 REMEDINAL-2 and S2013/MAE-
¤ Current address: Oliviculture Service, Madrid Institute for Research and Rural, Agrarian and Food Development, IMIDRA, Alcala´ de Henares, Spain * [email protected]
Abstract Regulation of agricultural pests managing their natural enemies represents an alternative to chemical pesticides. We assessed the potential of insectivorous birds as pest regulators in woody crops located in central Spain. A total of 417 nest boxes installed in five field study sites (one vineyard, two fruit orchards, and two olive groves) were monitored for use and breeding of insectivorous birds and other species for four consecutive years (2013–2016). At all field sites except the two olive groves, where birds never occupied the nest boxes, predation experiments were conducted with Greater wax moth (Galleria mellonella) sentinel caterpillars, and food consumption by birds was estimated. Nesting of insectivorous birds, chiefly Great tit (Parus major), and sparrows (Passer domesticus and P. montanus) increased over time, averaging 60% per field site in the vineyard and fruit orchards by the fourth year. Use of nest boxes by sparrows and by Garden dormouse (Eliomys quercinus) was high at the fruit orchards (70%) and the vineyard (30%), respectively. Micro-habitat characteristics (nest box level) and meso-habitat characteristics (patch level) strongly affected use of nest boxes and bird breeding (i.e. number of laid eggs and produced chicks) in different years. Distance to natural or semi-natural vegetation did not consistently affect bird breeding, nor did we see consistent evidence of competition between adjacent breeding birds. Predation rates of sentinel caterpillars were approximately one-third higher near boxes with nesting birds (31.51 ± 43.13%) than at paired distant areas without nest boxes (22.45% ± 38.58%). Food consumption by insectivorous birds per ha and breeding season were conservatively estimated to range from 0.02 kg in one fruit orchard to 0.15 kg in the vineyard. We conclude that installation of nest boxes in Mediterranean woody crops enhances populations of insectivorous birds that regulate pests, but that the effects are moderate and highly context-dependent.
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
1 / 19
Pest regulation by insectivorous birds
2719 REMEDINAL-3 to JMRB), Fundacio´n Biodiversidad (competitive call of 2012 granted to JMRB), and the ERANET-LAC programme (grant number PCIN-2016-150 to LC). DH-B received a scholarship from the Spanish Ministry of Education, Culture and Sports. We acknowledge support from Abadı´a Retuerta and HaciendasBio. Klaus Birkhofer and Eduardo Crisol-Martı´nez kindly reviewed the manuscript prior to submission. We are greatful to four anonymous reviewers who helped to improve this manuscript. Armando Chapin greatly improved English writing. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors declare the following interest: JMRB received a grant through the International Foundation for Ecosystem Restoration from the commercial company Abadı´a Retuerta. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction The need to maintain and enhance crop yield has led to an increase in agrochemicals use throughout the approximately 12% of the Earth’s ice-free land that is cultivated [1]. In particular, more than 2.4 billion tons of pesticides were used worldwide in 2007 [2]. Despite their advantages, pesticides can harm biodiversity [3, 4], including organisms beneficial to agricultural production itself (natural pest enemies, pollinators and decomposers [5]), and they can degrade the overall environment by polluting soil and water [6] and promoting soil erosion [7]. Pesticides can also threaten human health [8], and they can become ineffective when pests develop resistance [9]. An alternative to pesticide use is to regulate pests by maintaining or enhancing their natural enemies [10]. This strategy offers a more sustainable agricultural management because it can simultaneously promote biodiversity and ecosystem services other than crop production [11, 12, 13]. One third of the commercially available pesticides target arthropods [2], leading to the loss of an average of 15% of crops worldwide [14]. Crop insect pests have many natural enemies that could be exploited to manage the pest population as an alternative to pesticides; these enemies comprise several taxa, including bacteria, fungi, invertebrates, and vertebrates, which engage in diverse trophic relationships with pests, including predation, parasitism, parasitoidism, and disease. Previous research of pest enemies has focused mostly on microorganisms and invertebrates [15, 16, 17], while less attention has been paid to vertebrate species (e.g. birds) for crop pest regulation, particularly in temperate areas (but see [18, 19]). This research gap is important to address given that several studies have already suggested that insectivorous birds can be effective for controlling crop pests. For example, Great tits (Parus major) significantly reduced caterpillar damage by 19% and increased fruit yield by 66% in German commercial apple orchards [20]. The presence of nest boxes occupied by insectivorous birds in a Californian vineyard increased the average removal rate of Beet armyworms (Spodoptera exigua) by a factor of 3.5 [21]. In Costa Rica, birds reduced by half an infestation of Coffee berry borer beetles (Hypothenemus hampei) [22]. Excluding bats and birds from an ecosystem in Indonesia led to an increase in insect herbivore abundance and concomitant 31% decrease in final crop yield [23]. In the present study, we assessed the potential of insectivorous birds to regulate pests in Mediterranean woody crops. Woody crops cover nearly five million ha in Spain, accounting for 29.1% of cropland in the country [24]. They are the dominant vegetation in many agricultural landscapes in this and other regions of the world, and they are important habitats for species targeted by conservation efforts [25]. Woody crops occur in areas protected by the European Union Natura 2000 Network of natural protected areas [26]. They often demonstrate well sustainable agricultural practices that balance the requirement for agricultural production against the desire for wildlife conservation [27]. Here we studied vineyards (Vitis vinifera); orchards of Peach trees (Prunus persica), Prune trees (Prunus domestica), and Pear trees (Pyrus communis); and groves of Olive trees (Olea europaea). We installed nest boxes and conducted sentinel predation experiments for four consecutive years (2013–2016). Our specific objectives were to (1) evaluate nest box use by insectivorous birds and other species, (2) evaluate the factors that may explain breeding of insectivorous birds at nest boxes, (3) measure predation near nest boxes containing nesting birds and compare such predation with that in distant areas without nest boxes, and (4) estimate food consumption by insectivorous birds on selected farms. We hypothesized that (H1) birds’ use of nest boxes would increase with time, (H2) birds would favor habitat patches most suitable for breeding, (H3) more breeding would occur near patches of non-crop natural or
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
2 / 19
Pest regulation by insectivorous birds
semi-natural vegetation, and (H4) predation of sentinel caterpillars would be greater near nest boxes with nesting birds than in distant areas without nest boxes. As far as we know, this is the first systematic experimental work on pest regulation by insectivorous birds in various types of woody crops within the same study. The lessons learned here may be useful to farmers and conservation practitioners trying to move towards more wildlifefriendly farming.
Material & methods Study sites We studied the following five sites in Central Spain (Fig 1A): the 203-ha Abadı´a Retuerta vineyard (41.61˚ N, -4.41˚ W); the 62-ha Concejiles fruit orchard with Peach and Prune trees (38.83˚ N, -6.28˚ W); the 80-ha Chaparrito fruit orchard with Peach, Prune and Pear trees (38.94˚ N, -6.76˚ W); a 4.6-ha olive grove (38.81˚ N, -3.23˚ W); and a second olive grove of 1.9 ha (38.81˚ N, -3.36˚ W). No occupancy of insectivorous birds was detected at either olive tree grove throughout the four-year study, probably because they lay >2 km from suitable bird habitat and because the nest boxes were not covered by vegetation [15, 29–31]. Therefore these two sites were not considered further in this study. The vineyard was surrounded by Stone pine Pinus pinea plantations of different age (the border between the vineyard and pine plantations was 5,200-m long). The vineyard was also surrounded by natural Mediterranean vegetation dominated by Holm oak Quercus ilex subsp. ballota (the border was 3,200-m long), and there was a 1,350-m strip of riparian forest on the northwestern side (Fig 1B). The Concejiles fruit orchard was surrounded by rainfed cropland with sparse Holm oak Quercus ilex subsp. rotundifolia at a distance of 150 m on all sides except the northeastern side with presence of a 700-m strip of riparian vegetation and planted eucalyptus. Retama sphaerocarpa shrubland and Holm oak woodland lay, respectively, 380 and 700 m from the southwestern edge of this site (Fig 1C). The Chaparrito fruit orchard was surrounded by other fruit orchards and open rice fields; the northern edge of this site shared a 1,450-m border with a strip of degraded riparian vegetation (Fig 1D). Climate at all study sites is Mediterranean, with warm and dry summers and cold winters (S1 Table).
Setting and monitoring of nest boxes Since the crops at the study sites lacked appropriate cavities to serve as nesting sites for insectivorous birds, we installed 376 nest boxes in total to favor their breeding [32] at the three sites between January and mid-February in 2013. Nest boxes in the vineyard were made from wood; those in the fruit orchards were made from wood or pressed cardboard. The hole in the nest boxes had a diameter of 3.5 cm, widely used to favor nesting of the most common insectivorous bird species at the study sites, namely the Great tit (Parus major) and Blue tit (Cyanistes caeruleus). During the study, we noticed that other bird species and rodents, particularly the Garden dormouse (Eliomys quercinus), also used the installed nest boxes (see below). Thus, in February 2016 (the last year of the study), we left the nest boxes uncovered before the breeding season in order to reduce occupancy by Garden dormouse. All study sites included a treatment area with nest boxes as well as a distant area without nest boxes, and the two areas were separated by a buffer area (specific distances are reported below). Distant areas without nest boxes were used only for predation experiments (see below). In all treatment areas, the high number of nest boxes assured that availability of sites was not a limiting factor for bird breeding. Nevertheless, the separation between adjacent nest boxes was consistently 25 m because breeding individuals of small insectivorous birds regularly patrol small territories around their nest [33, 34]. In other respects, the layout and
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
3 / 19
Pest regulation by insectivorous birds
Fig 1. (A) Location of the five study sites in central Spain, namely the Abadı´a Retuerta vineyard, the Concejiles and Chaparrito fruit orchards and the Vistalegre and El Peral olive groves. (B) Distribution of the initial 164 nest boxes (black dots) in the Abadı´a Retuerta vineyard. (C) Distribution of the initial 101 nest boxes (black dots) in the Concejiles fruit orchard. (D) Distribution of the initial 111 nest boxes (black dots) in the Chaparrito fruit orchard. Light gray polygons indicate distant areas without nest boxes used for sentinel caterpillar experiments. Images were taken from [28] under a CC BY license, with permission from Google Maps/Google Earth (URL: https://www.google.es/intl/es/permissions/geoguidelines.html), original copyright 2015. The four panels are not identical to the original images, and are therefore for illustrative purposes only. https://doi.org/10.1371/journal.pone.0180702.g001
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
4 / 19
Pest regulation by insectivorous birds
characteristics of the “treatment area-buffer area-distant area” differed across study sites, reflecting differences in crop type, area, configuration, and surrounding vegetation. As a preliminary trial, we installed 63 nest boxes at the maximum vine height of 1.5 m inside the vineyard in february 2012. None of the boxes was occupied by birds, and 19 were used by wasps for nesting. Vine leaves sprout in late spring, so nest boxes were not covered when birds were exploring sites for nesting. Since the vineyard interior turned out to be a highly unsuitable habitat for insectivorous birds, we installed the nest boxes in clusters at the border between the pine plantations or Holm oak forest and the individual vineyard lots; the nest boxes were oriented with the entrance holes facing the vineyard. Within a cluster, nest boxes were placed in tree branches at heights >1.5 m above the ground. In each vineyard lot, the length of the cluster with nest boxes was equal to the length of the distant area without boxes; a 50-m buffer area lay between the cluster with nest boxes and the distant area without boxes. We initially placed 164 nest boxes at this study site (Fig 1B); a few were lost each year as follows: 3 in 2013, 9 in 2014, 12 in 2015, and 14 in 2016. The fruit orchards included individual fields planted with different tree species, and each field was sufficiently large to include at least one tree line with nest boxes and one distant tree line without nest boxes, with a buffer area at least 50-m wide separating the two tree lines. Nest boxes were placed in tree branches as high as possible, 1–2 m above the ground. At the Concejiles site, 101 nest boxes were installed initially (Fig 1C), and some were lost each year as follows: 3 in 2013, 9 in 2014, 18 in 2015, and 19 in 2016. At the Chaparrito site, 111 nest boxes were installed initially (Fig 1D), and some were lost each year: 3 in 2013, 11 in 2014, 16 in 2015, and 29 in 2016. We used a GPS to assess the exact position of every nest box at the two fruit orchards and the vineyard. Nest boxes were monitored for four consecutive years (2013–2016) from the start of the breeding season until the end of the summer in order to capture late recruitment (objectives 1–2). Each nest box was monitored four to five times per year except in 2014 and 2016, when the breeding season was quite short. In these years, nest boxes were monitored twice in the vineyard and three times in the fruit orchards. We judged this frequency of nest visits as adequate for assessing overall use of nest boxes and their potential for bird breeding, since we were interested in analyzing the use of nest boxes and not the phenology of breeding. We distinguished the following categories of nest-box use: 0 –empty nest box, 1 –initiated but unfinished bird nest, 2 –finished bird nest, 3 –laid eggs, 4 –hatched eggs, 5 –recruited fledglings, and 6 –presence of species other than birds. The number of recruited fledglings was determined on the basis of the number of individuals that were counted in the nest box at the last visit at the end of the breeding season, thus including second lays. In the autumn after the breeding season, nest boxes were cleaned and repaired or replaced if necessary.
Predation experiments Over the whole study period, we conducted sentinel predation experiments using larvae of Greater wax moth (Galleria mellonella). These experiments were designed to test whether predation rates differed between areas near nest boxes with breeding birds (“active” nest boxes sensu [21]) and distant areas without nest boxes (objective 3). We performed sentinel experiments on the day immediately after the nest boxes had been monitored at each site, except at the end-of-summer monitoring. We placed two samples of 10 caterpillars each, one closer and one farther away (specific distances are provided below) from each nest box containing laid eggs or chicks. Distance between adjacent caterpillars within a sample was 8 cm. In the vineyard, the sample closer to the active nest box lay 30% in years 2–4 (Fig 2A). Leaving nest boxes uncovered before the breeding season in the fourth year reduced occupancy by this rodent from 44% to 31% in the vineyard and from 24% to 0% in the Concejiles fruit orchard.
Explanatory factors of nest box use and breeding Spearman’s rank correlations of category of use across nest boxes between two consecutive years (H1) were positive and significant: p values for the three year-pairings were
Potential of pest regulation by insectivorous birds in Mediterranean woody crops Jose´ M. Rey Benayas1,2*, Jorge Meltzer2, Daniel de las Heras-Bravo1¤, Luis Cayuela2,3 1 Life Sciences Department-Ecology Unit, Alcala University, Alcala´ de Henares, Spain, 2 International Foundation for Ecosystem Restoration, Madrid, Spain, 3 Department of Biology, Geology, Physics and Inorganic Chemistry, Rey Juan Carlos University, Mo´stoles, Spain
a1111111111 a1111111111 a1111111111 a1111111111 a1111111111
OPEN ACCESS Citation: Rey Benayas JM, Meltzer J, de las HerasBravo D, Cayuela L (2017) Potential of pest regulation by insectivorous birds in Mediterranean woody crops. PLoS ONE 12(9): e0180702. https:// doi.org/10.1371/journal.pone.0180702 Editor: Tim A. Mousseau, University of South Carolina, UNITED STATES Received: January 31, 2017 Accepted: June 20, 2017 Published: September 6, 2017 Copyright: © 2017 Rey Benayas et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This research was funded by the Spanish Ministry of Economy and Competitivity (grant numbers CGL2010-18312 and CGL2014-53308-P to JMRB), the Spanish Ministry of Food, Agriculture and the Environment (calls to foundations numbers ARM/934/2011, AAA/2286/ 2012, AAA/1284/2014, and AAA/1770/2015 to JMRB), the government of Madrid (grant numbers S2009AMB-1783 REMEDINAL-2 and S2013/MAE-
¤ Current address: Oliviculture Service, Madrid Institute for Research and Rural, Agrarian and Food Development, IMIDRA, Alcala´ de Henares, Spain * [email protected]
Abstract Regulation of agricultural pests managing their natural enemies represents an alternative to chemical pesticides. We assessed the potential of insectivorous birds as pest regulators in woody crops located in central Spain. A total of 417 nest boxes installed in five field study sites (one vineyard, two fruit orchards, and two olive groves) were monitored for use and breeding of insectivorous birds and other species for four consecutive years (2013–2016). At all field sites except the two olive groves, where birds never occupied the nest boxes, predation experiments were conducted with Greater wax moth (Galleria mellonella) sentinel caterpillars, and food consumption by birds was estimated. Nesting of insectivorous birds, chiefly Great tit (Parus major), and sparrows (Passer domesticus and P. montanus) increased over time, averaging 60% per field site in the vineyard and fruit orchards by the fourth year. Use of nest boxes by sparrows and by Garden dormouse (Eliomys quercinus) was high at the fruit orchards (70%) and the vineyard (30%), respectively. Micro-habitat characteristics (nest box level) and meso-habitat characteristics (patch level) strongly affected use of nest boxes and bird breeding (i.e. number of laid eggs and produced chicks) in different years. Distance to natural or semi-natural vegetation did not consistently affect bird breeding, nor did we see consistent evidence of competition between adjacent breeding birds. Predation rates of sentinel caterpillars were approximately one-third higher near boxes with nesting birds (31.51 ± 43.13%) than at paired distant areas without nest boxes (22.45% ± 38.58%). Food consumption by insectivorous birds per ha and breeding season were conservatively estimated to range from 0.02 kg in one fruit orchard to 0.15 kg in the vineyard. We conclude that installation of nest boxes in Mediterranean woody crops enhances populations of insectivorous birds that regulate pests, but that the effects are moderate and highly context-dependent.
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
1 / 19
Pest regulation by insectivorous birds
2719 REMEDINAL-3 to JMRB), Fundacio´n Biodiversidad (competitive call of 2012 granted to JMRB), and the ERANET-LAC programme (grant number PCIN-2016-150 to LC). DH-B received a scholarship from the Spanish Ministry of Education, Culture and Sports. We acknowledge support from Abadı´a Retuerta and HaciendasBio. Klaus Birkhofer and Eduardo Crisol-Martı´nez kindly reviewed the manuscript prior to submission. We are greatful to four anonymous reviewers who helped to improve this manuscript. Armando Chapin greatly improved English writing. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors declare the following interest: JMRB received a grant through the International Foundation for Ecosystem Restoration from the commercial company Abadı´a Retuerta. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction The need to maintain and enhance crop yield has led to an increase in agrochemicals use throughout the approximately 12% of the Earth’s ice-free land that is cultivated [1]. In particular, more than 2.4 billion tons of pesticides were used worldwide in 2007 [2]. Despite their advantages, pesticides can harm biodiversity [3, 4], including organisms beneficial to agricultural production itself (natural pest enemies, pollinators and decomposers [5]), and they can degrade the overall environment by polluting soil and water [6] and promoting soil erosion [7]. Pesticides can also threaten human health [8], and they can become ineffective when pests develop resistance [9]. An alternative to pesticide use is to regulate pests by maintaining or enhancing their natural enemies [10]. This strategy offers a more sustainable agricultural management because it can simultaneously promote biodiversity and ecosystem services other than crop production [11, 12, 13]. One third of the commercially available pesticides target arthropods [2], leading to the loss of an average of 15% of crops worldwide [14]. Crop insect pests have many natural enemies that could be exploited to manage the pest population as an alternative to pesticides; these enemies comprise several taxa, including bacteria, fungi, invertebrates, and vertebrates, which engage in diverse trophic relationships with pests, including predation, parasitism, parasitoidism, and disease. Previous research of pest enemies has focused mostly on microorganisms and invertebrates [15, 16, 17], while less attention has been paid to vertebrate species (e.g. birds) for crop pest regulation, particularly in temperate areas (but see [18, 19]). This research gap is important to address given that several studies have already suggested that insectivorous birds can be effective for controlling crop pests. For example, Great tits (Parus major) significantly reduced caterpillar damage by 19% and increased fruit yield by 66% in German commercial apple orchards [20]. The presence of nest boxes occupied by insectivorous birds in a Californian vineyard increased the average removal rate of Beet armyworms (Spodoptera exigua) by a factor of 3.5 [21]. In Costa Rica, birds reduced by half an infestation of Coffee berry borer beetles (Hypothenemus hampei) [22]. Excluding bats and birds from an ecosystem in Indonesia led to an increase in insect herbivore abundance and concomitant 31% decrease in final crop yield [23]. In the present study, we assessed the potential of insectivorous birds to regulate pests in Mediterranean woody crops. Woody crops cover nearly five million ha in Spain, accounting for 29.1% of cropland in the country [24]. They are the dominant vegetation in many agricultural landscapes in this and other regions of the world, and they are important habitats for species targeted by conservation efforts [25]. Woody crops occur in areas protected by the European Union Natura 2000 Network of natural protected areas [26]. They often demonstrate well sustainable agricultural practices that balance the requirement for agricultural production against the desire for wildlife conservation [27]. Here we studied vineyards (Vitis vinifera); orchards of Peach trees (Prunus persica), Prune trees (Prunus domestica), and Pear trees (Pyrus communis); and groves of Olive trees (Olea europaea). We installed nest boxes and conducted sentinel predation experiments for four consecutive years (2013–2016). Our specific objectives were to (1) evaluate nest box use by insectivorous birds and other species, (2) evaluate the factors that may explain breeding of insectivorous birds at nest boxes, (3) measure predation near nest boxes containing nesting birds and compare such predation with that in distant areas without nest boxes, and (4) estimate food consumption by insectivorous birds on selected farms. We hypothesized that (H1) birds’ use of nest boxes would increase with time, (H2) birds would favor habitat patches most suitable for breeding, (H3) more breeding would occur near patches of non-crop natural or
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
2 / 19
Pest regulation by insectivorous birds
semi-natural vegetation, and (H4) predation of sentinel caterpillars would be greater near nest boxes with nesting birds than in distant areas without nest boxes. As far as we know, this is the first systematic experimental work on pest regulation by insectivorous birds in various types of woody crops within the same study. The lessons learned here may be useful to farmers and conservation practitioners trying to move towards more wildlifefriendly farming.
Material & methods Study sites We studied the following five sites in Central Spain (Fig 1A): the 203-ha Abadı´a Retuerta vineyard (41.61˚ N, -4.41˚ W); the 62-ha Concejiles fruit orchard with Peach and Prune trees (38.83˚ N, -6.28˚ W); the 80-ha Chaparrito fruit orchard with Peach, Prune and Pear trees (38.94˚ N, -6.76˚ W); a 4.6-ha olive grove (38.81˚ N, -3.23˚ W); and a second olive grove of 1.9 ha (38.81˚ N, -3.36˚ W). No occupancy of insectivorous birds was detected at either olive tree grove throughout the four-year study, probably because they lay >2 km from suitable bird habitat and because the nest boxes were not covered by vegetation [15, 29–31]. Therefore these two sites were not considered further in this study. The vineyard was surrounded by Stone pine Pinus pinea plantations of different age (the border between the vineyard and pine plantations was 5,200-m long). The vineyard was also surrounded by natural Mediterranean vegetation dominated by Holm oak Quercus ilex subsp. ballota (the border was 3,200-m long), and there was a 1,350-m strip of riparian forest on the northwestern side (Fig 1B). The Concejiles fruit orchard was surrounded by rainfed cropland with sparse Holm oak Quercus ilex subsp. rotundifolia at a distance of 150 m on all sides except the northeastern side with presence of a 700-m strip of riparian vegetation and planted eucalyptus. Retama sphaerocarpa shrubland and Holm oak woodland lay, respectively, 380 and 700 m from the southwestern edge of this site (Fig 1C). The Chaparrito fruit orchard was surrounded by other fruit orchards and open rice fields; the northern edge of this site shared a 1,450-m border with a strip of degraded riparian vegetation (Fig 1D). Climate at all study sites is Mediterranean, with warm and dry summers and cold winters (S1 Table).
Setting and monitoring of nest boxes Since the crops at the study sites lacked appropriate cavities to serve as nesting sites for insectivorous birds, we installed 376 nest boxes in total to favor their breeding [32] at the three sites between January and mid-February in 2013. Nest boxes in the vineyard were made from wood; those in the fruit orchards were made from wood or pressed cardboard. The hole in the nest boxes had a diameter of 3.5 cm, widely used to favor nesting of the most common insectivorous bird species at the study sites, namely the Great tit (Parus major) and Blue tit (Cyanistes caeruleus). During the study, we noticed that other bird species and rodents, particularly the Garden dormouse (Eliomys quercinus), also used the installed nest boxes (see below). Thus, in February 2016 (the last year of the study), we left the nest boxes uncovered before the breeding season in order to reduce occupancy by Garden dormouse. All study sites included a treatment area with nest boxes as well as a distant area without nest boxes, and the two areas were separated by a buffer area (specific distances are reported below). Distant areas without nest boxes were used only for predation experiments (see below). In all treatment areas, the high number of nest boxes assured that availability of sites was not a limiting factor for bird breeding. Nevertheless, the separation between adjacent nest boxes was consistently 25 m because breeding individuals of small insectivorous birds regularly patrol small territories around their nest [33, 34]. In other respects, the layout and
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
3 / 19
Pest regulation by insectivorous birds
Fig 1. (A) Location of the five study sites in central Spain, namely the Abadı´a Retuerta vineyard, the Concejiles and Chaparrito fruit orchards and the Vistalegre and El Peral olive groves. (B) Distribution of the initial 164 nest boxes (black dots) in the Abadı´a Retuerta vineyard. (C) Distribution of the initial 101 nest boxes (black dots) in the Concejiles fruit orchard. (D) Distribution of the initial 111 nest boxes (black dots) in the Chaparrito fruit orchard. Light gray polygons indicate distant areas without nest boxes used for sentinel caterpillar experiments. Images were taken from [28] under a CC BY license, with permission from Google Maps/Google Earth (URL: https://www.google.es/intl/es/permissions/geoguidelines.html), original copyright 2015. The four panels are not identical to the original images, and are therefore for illustrative purposes only. https://doi.org/10.1371/journal.pone.0180702.g001
PLOS ONE | https://doi.org/10.1371/journal.pone.0180702 September 6, 2017
4 / 19
Pest regulation by insectivorous birds
characteristics of the “treatment area-buffer area-distant area” differed across study sites, reflecting differences in crop type, area, configuration, and surrounding vegetation. As a preliminary trial, we installed 63 nest boxes at the maximum vine height of 1.5 m inside the vineyard in february 2012. None of the boxes was occupied by birds, and 19 were used by wasps for nesting. Vine leaves sprout in late spring, so nest boxes were not covered when birds were exploring sites for nesting. Since the vineyard interior turned out to be a highly unsuitable habitat for insectivorous birds, we installed the nest boxes in clusters at the border between the pine plantations or Holm oak forest and the individual vineyard lots; the nest boxes were oriented with the entrance holes facing the vineyard. Within a cluster, nest boxes were placed in tree branches at heights >1.5 m above the ground. In each vineyard lot, the length of the cluster with nest boxes was equal to the length of the distant area without boxes; a 50-m buffer area lay between the cluster with nest boxes and the distant area without boxes. We initially placed 164 nest boxes at this study site (Fig 1B); a few were lost each year as follows: 3 in 2013, 9 in 2014, 12 in 2015, and 14 in 2016. The fruit orchards included individual fields planted with different tree species, and each field was sufficiently large to include at least one tree line with nest boxes and one distant tree line without nest boxes, with a buffer area at least 50-m wide separating the two tree lines. Nest boxes were placed in tree branches as high as possible, 1–2 m above the ground. At the Concejiles site, 101 nest boxes were installed initially (Fig 1C), and some were lost each year as follows: 3 in 2013, 9 in 2014, 18 in 2015, and 19 in 2016. At the Chaparrito site, 111 nest boxes were installed initially (Fig 1D), and some were lost each year: 3 in 2013, 11 in 2014, 16 in 2015, and 29 in 2016. We used a GPS to assess the exact position of every nest box at the two fruit orchards and the vineyard. Nest boxes were monitored for four consecutive years (2013–2016) from the start of the breeding season until the end of the summer in order to capture late recruitment (objectives 1–2). Each nest box was monitored four to five times per year except in 2014 and 2016, when the breeding season was quite short. In these years, nest boxes were monitored twice in the vineyard and three times in the fruit orchards. We judged this frequency of nest visits as adequate for assessing overall use of nest boxes and their potential for bird breeding, since we were interested in analyzing the use of nest boxes and not the phenology of breeding. We distinguished the following categories of nest-box use: 0 –empty nest box, 1 –initiated but unfinished bird nest, 2 –finished bird nest, 3 –laid eggs, 4 –hatched eggs, 5 –recruited fledglings, and 6 –presence of species other than birds. The number of recruited fledglings was determined on the basis of the number of individuals that were counted in the nest box at the last visit at the end of the breeding season, thus including second lays. In the autumn after the breeding season, nest boxes were cleaned and repaired or replaced if necessary.
Predation experiments Over the whole study period, we conducted sentinel predation experiments using larvae of Greater wax moth (Galleria mellonella). These experiments were designed to test whether predation rates differed between areas near nest boxes with breeding birds (“active” nest boxes sensu [21]) and distant areas without nest boxes (objective 3). We performed sentinel experiments on the day immediately after the nest boxes had been monitored at each site, except at the end-of-summer monitoring. We placed two samples of 10 caterpillars each, one closer and one farther away (specific distances are provided below) from each nest box containing laid eggs or chicks. Distance between adjacent caterpillars within a sample was 8 cm. In the vineyard, the sample closer to the active nest box lay 30% in years 2–4 (Fig 2A). Leaving nest boxes uncovered before the breeding season in the fourth year reduced occupancy by this rodent from 44% to 31% in the vineyard and from 24% to 0% in the Concejiles fruit orchard.
Explanatory factors of nest box use and breeding Spearman’s rank correlations of category of use across nest boxes between two consecutive years (H1) were positive and significant: p values for the three year-pairings were
